[SNMP4J] UTF Encoding Support

[He haobo]

Hi Frank,

Thanks a lot. In my case, I will let user input a charset that will match
the charset defined in MIB specification first. For example, they might
input "gb2312". Then I will generate a new text just like new
String(variable.getValue(),"gb2312"); I guess this will meet my needs.
Thanks again.

Eternal


2009/1/6 <fock at agentpp.com>

> Hi,
>
> I have already written it: the MIB specification
> has to define which character set is used for
> a variable binding.
>
> To get the OctetString value never use toString().
> Instead use getValue()/toByteArray().
>
> Best regards,
> Frank
>
>
>
> Hi Frank,
>
> I took a look at the toString() method of OctetString that is like below.
>   public String toString() {
>     if (isPrintable()) {
>       return new String(value);
>     }
>     return toHexString();
>   }
>
>   public boolean isPrintable() {
>     for (int i=0; i<value.length; i++) {
>       char c = (char)value[i];
>       if ((Character.isISOControl(c) ||
>           ((value[i] & 0xFF) >= 0x80)) && (!Character.isWhitespace(c))) {
>         return false;
>       }
>     }
>     return true;
>   }
>
> So if the value byte array contains Chinese chars, it will not display the
> correct Chinese chars and will display the hexadecimal values. You know, in
> order to display the Chinese chars correctly, I have to do some encoding
> converting just like this.
> new String(((OctetString) vb[i].getVariable()).toByteArray(), "gb2312");
> After done that, the chinese chars will be correctly displayed. Therefore,
> I guess if the value byte array contains Japanese chars, I will also have to
> use a Japanese charset to reconstruct the text. Same process will be needed
> for other chars such as German, French and so on.
>
> My problem is that I am writing a snmp trap receive client and I can't know
> which chars the snmp trap sender(such as cisco router and some kinds of
> firewall) will use by advance. So in this case, I have to guess which
> charsets the snmp trap sender are using. Do you have any suggestion for this
> case? Thanks.
>
>
> Eternal
>
>
>
>
> 2009/1/2 Frank Fock <fock at agentpp.com>
>
>> Hi,
>>
>> OctetString does not "use" any encoding. As the name
>> suggests, it represents an octet string which is
>> a (binary) string of 8-bit bytes.
>>
>> If you convert an OctetString to a String, UTF-8 is
>> used by default (Java does that). You are responsible
>> to use the correct format as specified in the
>> corresponding MIB specification of the object.
>>
>> Regards,
>> Frank
>>
>>
>> karunakaran s.l wrote:
>>
>>> Hi All,
>>>
>>>            I want to know the snmp4j's  "OctetString" uses the UTF
>>> encoding
>>> or not.
>>>
>>> If it uses the UTF Encoding means what is encoding type (8 or 16). If not
>>> using UTF encoding means is any other way of using UTF Encoding for
>>> setting
>>> the MO Values of type OctetString. (Example: UTF encoding for
>>> sysLocation).
>>>
>>> Please clarify this as soon as possible.
>>>
>>>
>>> Thanks
>>> Karunakaran
>>> _______________________________________________
>>> SNMP4J mailing list
>>> SNMP4J at agentpp.org
>>> http://lists.agentpp.org/mailman/listinfo/snmp4j
>>>
>>
>> --
>> AGENT++
>> http://www.agentpp.com
>> http://www.mibexplorer.com
>> http://www.mibdesigner.com
>>
>>
>>
>> _______________________________________________
>> SNMP4J mailing list
>> SNMP4J at agentpp.org
>> http://lists.agentpp.org/mailman/listinfo/snmp4j
>>
>
>